In the intensive care of sick infants, the global oxygen reserve capacity is estimated by co-oximetry (co-ox) of blood sampled from central venous catheters. Introduction of a noninvasive alternative is desirable. Near infrared spectroscopy (NIRS) offers a technique for noninvasive bedside monitoring of tissue oxygen economy. We studied the relation between peripheral venous oxyhemoglobin saturation (SvO₂) estimated by venous occlusion and NIRS, and the central SvO₂ measured by co-ox of central venous blood. We report the high reproducibility of NIRS with a test-retest variation of only 2.51 ± 1.41%. After bias adjusting of NIRS SvO₂ values, a nice correlation (r = 0.96, p > 0.05) between NIRS measurements of peripheral SvO₂ and co-ox of central venous blood was found. The study indicates that NIRS is practical for monitoring relative changes in central venous saturation. This might be useful in the future clinical care of newborns.

Cerebrovascular reactivity to CO2 in clinical and experimental studies has been found to be impaired during increased intracranial pressure (ICP). However, from previous study results it has not been possible to estimate whether the impairment was caused by elevated ICP, or caused by decreased cerebral perfusion pressure (CPP). The current study was carried out in a group of unmanipulated control rats and in six investigation groups of six rats each: two groups with elevated ICP (30 and 50 mm Hg) and spontaneous arterial blood pressure (MABP), two groups with spontaneous ICP and arterial hypotension (77 and 64 mm Hg), and two groups with elevated ICP (30 and 50 mm Hg) and arterial hypertension (124 mm Hg). Intracranial hypertension was induced by continuous infusion of lactated Ringer's solution into the cisterna magna, arterial hypotension by controlled bleeding, and arterial hypertension by continuous administration of norepinephrine intravenously. Cerebral blood flow (CBF) was measured repetitively by the intraarterial 133Xe method at different levels of arterial PCO2. In each individual animal, CO2 reactivity was calculated from an exponential regression line obtained from the corresponding CBF/PaCO2 values. By plotting each individual value of CO2 reactivity against the corresponding CPP value from the seven investigation groups, CPP was significantly and directly related to CO2 reactivity of CBF (P < .001). No correlation was found by plotting CO2 reactivity values against the corresponding MABP values or the corresponding ICP values. Thus, the results show that CO2 reactivity is at least partially determined by CPP and that the impaired CO2 reactivity observed at intracranial hypertension and arterial hypotension may be caused by reduced CPP.

In subarachnoid hemorrhage (SAH) with cerebrovascular instability, hyperventilation may induce a risk of inducing or aggravating cerebral ischemia. We measured cerebral blood flow (CBF) and cerebral metabolic rates of oxygen (CMRO2), glucose (CMRglc), and lactate (CMRlac) at different PaCO2 levels after experimental SAH in rats (injection of 0.07 mL of autologous blood into the cisterna magna). Four groups of Sprague-Dawley male rats were studied at predetermined PaCO2 levels: group A: normocapnia (5.01-5.66 kPa [38.0-42.0 mm Hg]); group B: slight hyperventilation (4.34-5.00 kPa [32.5-37.5 mm Hg]); group C: moderate hyperventilation (3.67-4.33 kPa [27.5-32.4 mm Hg]); group D: profound hyperventilation (3.00-3.66 kPa [22.5-27.4 mm Hg]). Each of the four groups included eight rats with SAH and eight sham-operated controls. CBF was determined by the intracarotid Xe method; CMRo2, CMRglc, and CMRlac were obtained by cerebral arteriovenous differences. In both SAH rats and controls, hyperventilation decreased CBF in proportion to the decrement in PaCO2 without affecting either CMRO2, CMRglc, or CMRlac. In groups C and D, CBF decreased by 20%-35%, but CMRs were maintained by a compensatory increase in oxygen extraction fraction (OEF). The results show that even profound hyperventilation in this model of SAH is associated with an adequate increase in OEF so that CMRs of oxygen, glucose, and lactate remain similar to levels observed in normocapnic conditions.

Using a ventriculocisternal perfusion method, the effects of cerebrospinal fluid (CSF) acidity of nonrespiratory origin on cerebral blood flow (CBF) and autoregulation of CBF were investigated. Three groups (six rats each) were studied: one group of sham operated rats, one control group with ventriculocisternal perfusion at normal pH (mean inflow pH +/- SD, 7.42 +/- 0.02), and one experimental group with ventriculocisternal perfusion at low pH (mean inflow pH +/- SD, 6.81 +/- 0.01). CBF was measured by the intracarotid xenon 133 method. Autoregulation was studied by repetitive measurements of CBF during an initial increase and then stepwise reduction of mean arterial blood pressure (MABP). No difference in CBF was found between sham operated and control rats with unperturbed pH (mean cisternal outflow pH +/- SD, 7.42 +/- 0.03) of CSF), and autoregulation was intact in both groups. In the experimental group, the mean CBF +/- SD was increased by 58%, from 127 +/- 33 mL/(100 g.min) before ventriculocisternal perfusion to 201 +/- 54 mL/(100 g.min) (P <.00001) during perfusion with acid CSF (mean cisternal outflow pH +/- SD, 7.23 +/- 0.04). In this group, the relationship between CBF and MABP was linear, thus indicating disrupted autoregulation. In conclusion, CSF acidity significantly increases CBF and impairs autoregulation of CBF.